FinanceBenchmark: Financial AI Leaderboard 2026

FinanceBenchmark provides independent rankings and a composite score for leading Artificial Intelligence models, including Anthropic Claude, OpenAI GPT, and Google Gemini, on complex financial reasoning and forensic accounting tasks. Enable JavaScript to view the interactive, continuously updated leaderboard.

2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 • 2026-10-20UPDATEPlatform updated: 52 models tracked across 42 independent benchmarks including CorpFin • 2026-10-20BENCHCorpFin added to benchmark suite — long credit agreement comprehension evaluations live • 2026-10-19MODELClaude Sonnet 5.5 and Opus 5.5 achieve 75.9% and 75.8% on Excel Modeling Benchmark v1.1 • 2026-10-19MODELTax Agent Bench updated: Claude Fable 5.1 (49.2%) and Gemini 4 Argon (44.6%) lead corporate tax research • 2026-10-18MODELGemini 4 Argon claims #1 rank on Finance Agent (v2) with 65.4% • 2026-10-13MODELGoogle launches Gemini 4 Argon with 1M-token output limit and native multimodal reasoning • 2026-10-06MODELAnthropic completes Claude 5.5 generation with Claude Haiku 5.5, Opus 5.5, and Sonnet 5.5 •

Tier 2numericalforensic

MortgageTax Benchmark

Evaluates financial calculations and regulatory tax computations on complex mortgage agreements.

Variable tasks

21 models evaluated

Updated 2026

View Source →Submit your score Methodology

72.1%

Top Score

Top model

Claude Opus 5

Model Rankings

financebenchmark.ai · MortgageTax

64.2%

🥇

AnthropicClaude Opus 4.770.3%

🥈

OpenAIGPT-5.568.8%

🥉

AnthropicClaude Opus 572.1%

4

AnthropicClaude Sonnet 570.0%

5

OpenAIGPT-5.6 Sol67.3%

6

xAIGrok 4.664.2%

7

Moonshot AIKimi K3OPEN66.3%

8

MiniMaxMiniMax M3OPEN68.4%

9

AnthropicClaude Fable 568.9%

10

GoogleGemini 3.5 Flash Lite68.7%

11

GoogleGemini 3.6 Flash67.6%

12

OpenAIGPT-5.6 Luna67.3%

13

OpenAIGPT-5.6 Terra67.3%

14

xAIGrok 4.561.8%

15

Thinking MachinesInklingOPEN64.0%

16

Thinking MachinesInkling SmallOPEN62.3%

17

MetaMuse Spark 1.166.1%

18

MetaMuse Spark 1.265.4%

19

AlibabaQwen 3.7 Plus66.2%

20

AlibabaQwen 3.8 MaxOPEN64.0%

21

AnthropicClaude Fable 5.170.8%

Full Results Table

#	Model	Overall	Source	Date
🥇	Claude Opus 5 Anthropic	72.1	VALS (2026)	2026-09
🥈	Claude Fable 5.1 Anthropic	70.8	VALS (2026)	2026-09
🥉	Claude Opus 4.7 Anthropic	70.3	VALS (2026)	2026-09
4	Claude Sonnet 5 Anthropic	70.0	VALS (2026)	2026-09
5	Claude Fable 5 Anthropic	68.9	VALS (2026)	2026-09
6	GPT-5.5 OpenAI	68.8	VALS (2026)	2026-09
7	Gemini 3.5 Flash Lite Google	68.7	VALS (2026)	2026-09
8	MiniMax M3OPEN MiniMax	68.4	VALS (2026)	2026-09
9	Gemini 3.6 Flash Google	67.6	VALS (2026)	2026-07
10	GPT-5.6 Terra OpenAI	67.3	VALS (2026)	2026-07
11	GPT-5.6 Sol OpenAI	67.3	VALS (2026)	2026-07
12	GPT-5.6 Luna OpenAI	67.3	VALS (2026)	2026-07
13	Kimi K3OPEN Moonshot AI	66.3	VALS (2026)	2026-07
14	Qwen 3.7 Plus Alibaba	66.2	VALS (2026)	2026-06
15	Muse Spark 1.1 Meta	66.1	VALS (2026)	2026-07
16	Muse Spark 1.2 Meta	65.4	VALS (2026)	2026-08
17	Grok 4.6 xAI	64.2	VALS (2026)	2026-08
18	InklingOPEN Thinking Machines	64.0	VALS (2026)	2026-07
19	Qwen 3.8 MaxOPEN Alibaba	64.0	VALS (2026)	2026-08
20	Inkling SmallOPEN Thinking Machines	62.3	VALS (2026)	2026-07
21	Grok 4.5 xAI	61.8	VALS (2026)	2026-07